Search papers, labs, and topics across Lattice.
4
0
8
Smooth solutions to generalized SQG equations persist for nearby singularity parameters, bridging the gap between standard SQG and more singular active scalar models.
Fine-grained MoE designs can outperform dense vision encoders while dramatically reducing latency, challenging the status quo in image and video understanding.
BluTrain outpaces PyTorch in both speed and memory efficiency, setting a new standard for AI training frameworks.
Speech LLMs can now better understand your emotions: a new RL approach boosts paralinguistic understanding by 8-12% over state-of-the-art models.
